Skip to main content



Acute leukemia is an accumulation of progressively increasing numbers of immature precursor cells that fail to differentiate. Transcription factors are critical for normal hematopoietic differentiation, and a common genetic mechanism underlying the block in differentiation is dysregulation of transcription. Understanding regulation of these transcription factors is therefore relevant to both normal hematopoiesis and leukemia. The transcription factor PU.1 is essential for normal myeloid development regulating nearly every myeloid gene and is expressed early during hematopoietic differentiation. Downregulation of PU.1 expression leads to development of acute myeloid leukemia (AML), both in mouse experimental systems and studies in human cell lines. To achieve optimal expression of the PU.1 gene at different stages of hematopoiesis, multiple interactions are required between regulatory elements to form fine chromatin architecture. While significant advances have been made in understanding which PU.1 locus chromatin states are active or silent, nothing is known how these states are propagated and maintained. We have demonstrated a novel mechanism of gene regulation through coordinated expression of the PU.1 target gene and its regulatory natural noncoding antisense transcripts. In this study we investigated mechanisms in formation of alternant chromatin states and shed light on the interplay between well established epigenetic marks (such as chromatin compaction, histone modification, and DNA methylation) and the emerging class of new epigenetic marks: Noncoding antisense RNAs. The overall objective of this proposal is to elucidate the roles of noncoding antisense RNAs corresponding to the PU.1 gene locus in the regulation of PU.1 sense expression and hematopoietic stem cell differentiation.

We performed real time PCR analysis to quantify PU.1 mRNA and antisense transcription during blood cell differentiation. We further analyzed the methylation pattern of the PU.1 proximal promoter and determined chromosomal configuration of PU.1 gene locus.
We investigated the functional effect of noncoding antisense RNAs ablation on the formation of the active chromatin configuration and, ultimately, on the expression of the PU.1 gene in tissue culture and in vivo. We also evaluated the anti-leukemic effect of PU.1 antisense ablation in xenotransplantation models in vivo using cell lines and primary patient samples. Sense and antisense gene reporter assays were performed to evaluate the effect of targeting the fusion oncoproteins CBFb-MYH11 (inv16) and AML1-ETO (t8;21) on antisense promoter activity. Chromosome folding and methylation pattern of PU.1 proximal promoter was evaluated upon targeting the fusion oncoproteins CBFb-MYH11 (inv16) and AML1-ETO (t8;21).

We observed a strong inverse correlation of PU.1 mRNA and antisense transcription during blood cell differentiation with high PU.1 mRNA and low antisense levels in myeloid cells and low PU.1 mRNA and high antisense levels in T-cells. Transcription was accompanied by a specific promoter methylation pattern and chromosomal configuration.
We found pronounced functional effects of noncoding antisense RNAs ablation on the chromatin structure and the expression of the gene in vitro and in vivo. PU.1 antisense knockdown induced myeloid differentiation and apoptosis in leukemic cells of patients with the cytogenetic abnormalities inv.16 and t8;21 in vitro and prolonged leukemia development and survival in mice transplanted with these leukemic cells. We identified a comprehensive mechanism of core binding factor leukemias (CBFb-MYH11 (inv16) and AML1-ETO (t8;21)): Core binding factor fusions induce a T-cell like chromosomal state with increased antisense transcription resulting in methylation of the proximal promoter.

The achievement of this goal led to critical advances in the way we think about regulation of transcription and leukemogenesis via oncofusions.